Efficient Pseudotyping of Different Retroviral Vectors Using a Novel

Brief Report Efﬁcient Pseudotyping of Different Retroviral Vectors Using a Novel, Codon-Optimized Gene for Chimeric GALV Envelope

Manuela Mirow 1, Lea Isabell Schwarze 1,2, Boris Fehse 1,2,* and Kristoffer Riecken 1,*

1 Research Department Cell and Gene Therapy, Department of Stem Cell Transplantation, University Medical Center Hamburg-Eppendorf (UKE), 20246 Hamburg, Germany; [email protected] (M.M.); [email protected] (L.I.S.) 2 German Center for Infection Research (DZIF), Partner Site Hamburg-Lübeck-Borstel-Riems, 20246 Hamburg, Germany * Correspondence: [email protected] (B.F.); [email protected] (K.R.); Tel.: +49-40-7410-55518 (B.F.); +49-40-7410-52705 (K.R.)

Abstract: The Gibbon Ape Leukemia Virus envelope protein (GALV-Env) mediates efﬁcient transduction of human cells, particularly primary B and T lymphocytes, and is therefore of great interest in gene therapy. Using internal domains from murine leukemia viruses (MLV), chimeric GALV-Env

proteins such as GALV-C4070A were derived, which allow pseudotyping of lentiviral vectors. In order to improve expression efﬁciency and vector titers, we developed a codon-optimized (co) variant of

GALV-C4070A (coGALV-Env). We found that coGALV-Env mediated efficient pseudotyping not only of γ-retroviral and lentiviral vectors, but also α-retroviral vectors. The obtained titers on HEK293T cells were equal to those with the classical GALV-Env, whereas the required plasmid amounts for transient vector production were significantly lower, namely, 20 ng coGALV-Env plasmid per Citation: Mirow, M.; Schwarze, L.I.; 106 293T producer cells. Importantly, coGALV-Env-pseudotyped γ- and α-retroviral, as well as Fehse, B.; Riecken, K. Efficient lentiviral vectors, mediated efficient transduction of primary human T cells. We propose that the Pseudotyping of Different Retroviral Vectors Using a Novel, novel chimeric coGALV-Env gene will be very useful for the efficient production of high-titer vector Codon-Optimized Gene for Chimeric preparations, e.g., to equip human T cells with novel specificities using transgenic TCRs or CARs. GALV Envelope. Viruses 2021, 13, The considerably lower amount of plasmid needed might also result in a significant cost advantage 1471. https://doi.org/ for good manufacturing practice (GMP) vector production based on transient transfection. 10.3390/v13081471 Keywords: retroviral vectors; adoptive immunotherapy; chimeric antigen receptor (CAR); T-cell Academic Editors: Irene Gil-Farina receptor (TCR); gene therapy; pseudotyping; GMP and Manfred Schmidt

Received: 4 June 2021 Accepted: 25 July 2021 1. Introduction Published: 27 July 2021 Integration into the host cells’ genome is an obligatory step in the life cycle of retroviruses. Based thereon, vectors derived from γ-retro- as well as (later) lenti- and Publisher’s Note: MDPI stays neutral α-retroviruses have become important tools for gene-therapy applications that require with regard to jurisdictional claims in published maps and institutional affil- permanent expression in replicating cell populations, such as hematopoietic stem cells or T iations. lymphocytes [1–3]. The cell-specificity of retroviruses and vectors derived thereof is largely determined by their envelope (Env) proteins that mediate binding to cellular surface proteins also referred to as “viral receptors” (e.g., CD4 for HIV-Env). Early on in retrovirology, it was shown that the natural Env protein of retroviral particles could be replaced by suitable Copyright: © 2021 by the authors. equivalents (glycoproteins) derived from other viruses [3–6]. Such exchanges have been Licensee MDPI, Basel, Switzerland. subsumed under the term pseudotyping and today include not only the use of naturally This article is an open access article distributed under the terms and occurring, but also engineered viral envelopes [7]. Depending on the vector, exchange conditions of the Creative Commons of the original Env has several potential advantages, e.g., broader or narrower range of Attribution (CC BY) license (https:// target cells, lower toxicity, and higher stability allowing enrichment of vector particles creativecommons.org/licenses/by/ by ultracentrifugation [6]. Therefore, pseudotyping, particularly of lentiviral vectors, 4.0/). e.g., with the VSV-G protein, has become an indispensable tool to ensure efficient vector

Viruses 2021, 13, 1471. https://doi.org/10.3390/v13081471 https://www.mdpi.com/journal/viruses Viruses 2021, 13, 1471 2 of 11

production and achieve high-level gene transfer to desired cell populations [7]. On the other hand, the VSV-G protein has certain limitations, not the least its relatively high cytotoxicity, impeding its stable expression in permanent producer cell lines. Therefore, the pseudotyping potential of Env proteins derived from other retroviruses has been investigated, including those derived from murine retroviruses (MuLV10A1) [8], nonhuman primate viruses (Gibbon Ape Leukemia Virus (GALV) [4,9], Baboon retroviral envelope (BaEV) glycoprotein [10], and Feline Leukemia Virus (RD114) [5]. Already by the 1990s, GALV-Env-pseudotyped γ-retroviral vectors were shown to mediate efficient transduction of human primary T cells even with non-concentrated γ- retroviral vectors stocks [9,11]. In contrast, pseudotyping of lentiviral vectors with original GALV-Env was inefficient. This limitation was overcome by fusing the external, cell- binding subunit of GALV-Env to the transmembrane C-terminus of the amphotropic MLV 4070A thus generating the chimeric (GALV-C4070A) envelope [12,13]. Indeed, we previously applied lentiviral vectors pseudotyped with the chimeric GALV-C4070A Env for efficient transduction of primary T and B lymphocytes [14,15]. The recent breakthroughs in adoptive immunotherapy with gene-modified T cells expressing chimeric antigen receptors (CARs) [16] have further highlighted the need for methods that efficiently transduce large numbers of T cells in accordance with good manufacturing practice (GMP) regulations. Currently, manufacture of most viral CAR vectors is based on transient transfection of so-called producer cell lines with separate packaging plasmids encoding all necessary viral proteins as well as the vector’s genome. While sufficiently effective, this process is relatively variable and expensive. The need for large amounts of highly purified plasmids is one of the factors contributing to these high costs [17]. The ambiguity of the genetic code has resulted in the variable use of the 61 individual codons by different organisms. Indeed, a codon preferably used to encode a defined amino acid in a given organism might rarely be present in another. This observation has led to the concept of codon optimization [18] which represents a method that uses synonymous codon changes to increase protein production in defined target cells. The approach is based on the assumptions that rare codons are limiting the translation rate, and that changing synonymous codons does not affect the structure and function of a protein. While viruses generally adapt to the preferred codon usage of their host organisms during evolution, this adaptation is often hampered by their own biology. For example, the small genome of retroviruses requires the use of interlaced open-reading frames (ORFs) and the overlap of ORFs with indispensable structural elements both limiting the usage of perfectly adapted codons in the virus genome. In addition, the use of non-optimal codons might reflect another layer of regulation to ensure correct stoichiometry of the different viral proteins, which in case of retroviruses are all expressed by a single promoter. However, separation of the necessary genes on individual packaging plasmids has opened the possibility to codon- optimize cDNAs for improved vector packaging [18,19]. In contrast, codon optimization has been less frequently used to codon-optimize Env proteins. A recent paper even reported functional impairment of the RD114TR envelope glycoprotein after “codon optimization” due to incorrect protein processing [20]. Here, we describe the development and testing of a codon-optimized (co) gene variant of the chimeric GALV-C4070A-env. The new coGALV-Env was used to pseudotype lentiviral as well as γ- and α-retroviral vectors. We investigated the impact of codon optimization on vector titers and transduction efficiencies in cultured cell lines (HEK293, K562) and in primary human T cells. Our data demonstrate substantially improved translation efficiency with coGALV-env supporting its use for the production of all three vector types (lentiviral, γ- and α-retroviral vectors). Viruses 2021, 13, 1471 3 of 11

2. Materials and Methods 2.1. Plasmids and Codon Optimization The plasmid phCMV-GALV-C4070A encoding the chimeric GALV-C4070A-env [14] was a kind gift from Carol Stocking. The corresponding codon-optimized gene version coGALV-env was designed with GeneOptimizer software and synthesized by a contract manufacturer (Thermo Fisher Scientiﬁc, Waltham, MA, USA). A comparison of the complete original and the optimized sequences is provided in Supplementary Figure S1; vector maps are depicted in Supplementary Figure S2. Large amounts of plasmids were generated in E. coli TOP10 using standard recombinant-DNA techniques (QIAquick Spin Maxiprep Kit, QIAGEN, Hilden, Germany). The new plasmid containing the codon-optimized coGALV-env has been deposited at Addgene (#172217).

2.2. Generation of Viral Particles The used lentiviral vector LeGO-mOrange2 (Addgene, #85212) was described previously [21]. It was produced following our established protocol [22], termed “standard scale”, in 10-cm cell culture dishes (a detailed protocol is available online at http: //www.LentiGO-Vectors.de; assessed 25 July 2021). HEK293T cells (CRL-11268), obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA), were used for LV production and titration. For determination of the optimal amount of env-plasmid, virus production was conducted on a small scale, deviating from the standard protocol as follows: 8 × 105 293T cells were seeded per well of a 6-well plate in 1 mL DMEM medium. Consequently, 1/6 of the plasmid amounts were used for the DNA/CaCl2 mix. Instead of adding DNA/CaCl2 to HBS dropwise while air was blown into the HBS-buffer, the DNA/CaCl2 was mixed with the HBS buffer immediately in small-scale production. Viral particles were harvested after 12, 24, and 36 h. Harvest was performed by removing cellular debris via ﬁltration (0.45 µm), and the viral supernatants were stored at −80 ◦C. The γ-retroviral vector pRSF91.GFP.pre* [23] as well as both α-retroviral vectors (pAlpha.SIN(noTATA).EFS.EGFP.wPRE and pAlpha.SIN(noTATA).MPSV.EGFP.wPRE [24]) were kindly provided by Axel Schambach (MHH Hannover, Germany).

2.3. Titration of Vector Supernatants on 293T Cells 5 × 104 293T cells in 500 µL DMEM medium were seeded in a 24-well plate and incubated at 37 ◦C for 2–5 h until the cells were attached. Polybrene (Sigma-Aldrich, St. Louis, MO, USA) was added at a concentration of 8 µg/mL to each well, before 0.5 to 100 µL of the viral supernatant were given to the cells. Afterwards, the plates were centrifuged for 1 h at 25 ◦C and 1000× g (spin-inoculation). After 3 days of incubation at 37 ◦C, transduction efﬁciencies were analyzed by ﬂow cytometry.

2.4. Transduction of Myeloid K562 Cells and Primary Human T Lymphocytes Peripheral blood mononuclear cells (PBMCs) isolated from buffy coats (from healthy blood donors, kindly provided by the Institute of Transfusion Medicine, UKE, after informed consent) were stimulated with CD3/CD28 beads and IL-2 to selectively expand T cells. To this end, frozen PBMCs were thawed, washed, taken up in TexMACS medium with autologous serum, and subsequently seeded into a 6-well plate with 6 × 106 cells in 4 mL per well. Further, 200 U/mL IL-2 (Proleukin 100,000 U/mL, Novartis, Basel, Schweiz) and CD3/CD28 beads (T Cell TransAct, Miltenyi Biotec, Bergisch Gladbach, Germany) at a 1:100 ratio were added [11,25]. After an incubation time of 3 days, 2 × 105 primary T cells per well in 300 µL TexMACS medium with autologous serum were seeded in a 24-well plate, where only the 8 inner wells of the plate were used (in our experience, the outer wells sometimes were not even, leading to cells being pelleted at the border of a well after centrifugation, potentially interfering with transduction efﬁciencies). A total of 8 µg/mL of polybrene was added to each well, before 5 µL to 100 µL of the viral-particle containing supernatant were given to the cells. Subsequently, the plates were centrifuged for 1 h Viruses 2021, 13, 1471 4 of 11

at 25 ◦C and 1000× g. After 3 days of incubation at 37 ◦C, the cells were measured by ﬂow cytometry. Transduction of K562 cells (ATCC: CCL-243) was tested using 5 × 104 cells in 500 µL RPMI medium per well in a 24-well plate, where only the 8 inner wells of the plate were used. A total of 8 µg/mL of polybrene was added to each well before 1 µL to 50 µL of the vector supernatants were given to the cells. Afterwards, the plates were centrifuged, incubated, and analyzed in the same way as the primary T cells.

2.5. Statistical Analysis If not indicated otherwise, the data shown represent means and standard deviations of three independent experiments. Statistical significance was determined using Student’s t-test.

3. Results 3.1. Impact of Codon Optimization on Lentiviral Vector Titers With the aim of improving the expression strength of the envelope plasmid phCMV- GALV-C4070A encoding the chimeric GALV-C4070Aenv [14], a codon-optimized version coGALV-env was designed. A comparison of the complete original and the optimized sequences is provided in Supplementary Figure S1; plasmid maps are depicted in Supple- mentary Figure S2. We first assessed a potential influence of codon optimization on the titer of GALV-Env- pseudotyped lentiviral vectors (LVVs). To do so, we produced LeGO-mOrange2 vectors in 293T cells following our standard protocol (see Material and Methods), where the VSV-G protein was replaced by GALV-Env. Based on our own experience and data published by others [12,13,26,27], between 400 and 1000 ng GALV-env plasmid is required per 1 Mio cells to reproducibly ensure high-titer LVV production. In line with that, in preliminary experiments that were aimed at defining the optimal concentration range for the two plasmids, we determined peak titers (up to >4 × 106/mL) with approximately 550 ng GALV-env plasmid, whereas low amounts (≤100 ng/Mio cells) in some cases resulted in titers substantially below 105/mL. In contrast, for coGALV-Env peak titers of >3 × 106/mL could be obtained with approximately 10 ng plasmid per 1 Mio cells in this screening (Supplementary Figure S3). In the following, we tested different amounts of plasmids encoding either non-codon- optimized or coGALV-env in parallel experiments with identical passages of producer cells. In order to facilitate direct comparison, we focused on the respective optimal concentration ranges determined for the two different plasmids in the preliminary experiments. Super- natants were collected after approximately 12, 24, and 36 h; titers were assessed on 293T cells based on mOrange2 expression by flow cytometry. We obtained very similar results for all different time points (Figure1a). Therefore, here, we will focus on the 24 h data for the sake of clarity. The results of eight independent vector productions (all performed in triplicates for each concentration indicated) are displayed in Figure1. Based on the data shown in Figure1, we determined the optimal plasmid amounts for lentiviral vector production to be 640 ng/106 293T cells for standard GALV-env and 20 ng/106 293T cells for coGALV-env. Next, we directly compared these amounts of plasmids in a standard-scale vector production (in triplicates). As shown in Figure2, we indeed obtained identical titers with the >30 times lower amount of plasmid encoding the coGALV-Env. VirusesViruses2021 2021, ,13 13, 1471, 1471 55 of 1111

Viruses 2021, 13, 1471 5 of 11

(a) (b) (c) (d) (a) (b) (c) (d)

(e) (f) (g) (h) (e) (f) (g) (h)

FigureFigure 1.1.Comparative Comparative titrationtitration ofof lentivirallentiviral LeGO-mOrange2LeGO-mOrange2 vectorsvectors producedproduced withwith standardstandard vs.vs. codon-optimizedcodon-optimized (co)(co) FigureGALV-env.GALV-env. 1. Comparative ( a(a)) Much Much lowertitration lower amounts amounts of lentiviral ofof coGALV-envcoGALV-env LeGO-mOrange2 as as comparedcompared vectors to toproduced GALV-envGALV-env with were were standard required required vs. to tocodon-optimized obtainobtain maximalmaximal (co)titers.titers. GALV-env.SupernatantsSupernatants (a) Much werewere harvestedlowerharvested amounts 12,12, 24,24, of andand coGALV-env 3636 hh after after transfection transfectionas compared ofof to 293T293T GALV-env cells. cells. AsAs were evident,evident, required maximalmaximal to obtain titerstiters maximal werewere ininall alltiters. cases cases Supernatantsobtainedobtained withwith were >20>20 harvested timestimes lowerlower 12, amountsamounts24, and 36 ofof h coGALV-envcoGALV-env after transfection asas comparedcompared of 293T tocells.to GALV-env;GALV-env; As evident, ( b(b– –hmaximalh)) titration titration titers experiments experiments were in all for for cases seven seven obtainedadditional,additional, with independent independent>20 times lower vector vector amounts productions producti of coGALV-envons (only (only the 24the as h data24compared h pointdata is topoint shown). GALV-env; is shown). Means (b– and hMeans) titration standard and experiments deviationsstandard deviations of for triplicates seven of additional,aretriplicates shown. independent are Small shown. error Small barsvector are error producti part bars hidden areons part behind(only hidden the data 24 behind point h data symbols.data point point is symbols.shown). Means and standard deviations of triplicates are shown. Small error bars are part hidden behind data point symbols.

(a) (b) (c) (d) (a) (b) (c) (d) Figure 2. Titers of lentiviral LeGO-mOrange2 vectors produced in standard-scale with GALV-env and coGALV-env. FigureFigureGALV-env-vectors 2. 2.TitersTiters of oflentiviral lentiviralwere produced LeGO-mOrange2 LeGO-mOrange2 in triplicates vectors vectors with produced 640 produced ng ofin plasmid/10 instandard-scale standard-scale6 cells, with coGALVenv-vectors with GALV-env GALV-env and and coGALV-env. with coGALV-env. 20 ng of GALV-env-vectorsGALV-env-vectorsplasmid/106 cells. were (a were–c) producedExemplary produced in dotplots in triplicates triplicates of transduced with with 640 640 ng primary ng of of plasmid/10 plasmid/10 human T6 cells,cells6 cells, forcoGALVenv-vectors coGALVenv-vectorsthree different amounts with with 20(10, 20 ng 30, ng of 100 of plasmid/10plasmid/10µL) of vector-containing6 cells.6 cells. (a–c ()a Exemplary–c) Exemplarysupernatant dotplots dotplots prod ofuced transduced of transducedwith GALV-env primary primary humanor coGALV-env. human T cells T for cells Clthree forear-cut threedifferent separation different amounts of amounts transduced (10, 30, (10, 100 30,vs. µL)100non-transduced ofµ vector-containingL) of vector-containing cells even supernatant in supernatant demanding produced produced cells with facilitates GALV-env with GALV-envunequivocal or coGALV-env. or determination coGALV-env. Clear-cut Clear-cutof transduction separation separation of rates transduced of indicated; transduced vs. ( d) Means and standard deviations of titer determination performed in triplicates. No significant difference in titers as non-transducedvs. non-transduced cells even cells evenin demanding in demanding cells facilitates cells facilitates unequivocal unequivocal determination determination of transduction of transduction rates ratesindicated; indicated; (d) Meansmeasured and standardon 293T cellsdeviations was seen of bytiter t-test. determination FSC = forward perfor scatter,med AUin triplicates.= arbitrary unitsNo significant of area signal. difference in titers as (d) Means and standard deviations of titer determination performed in triplicates. No signiﬁcant difference in titers as measured on 293T cells was seen by t-test. FSC = forward scatter, AU = arbitrary units of area signal. measured on 293T cells was seen by t-test. FSC = forward scatter, AU = arbitrary units of area signal. 3.2. Titer of GALV-Env vs. coGALV-Env Pseudotyped γ-Retroviral (γ-RVVs) and α-Retroviral 3.2.Vectors Titer of (α GALV-Env-RVVs) vs. coGALV-Env Pseudotyped γ-Retroviral (γ-RVVs) and α-Retroviral Vectors3.2. Titer γ(α-retroviral-RVVs) of GALV-Env vector vs. systems coGALV-Env are frequently Pseudotyped usedγ-Retroviral in gene therapy, (γ-RVVs) particularly and α-Retroviral for T- Vectorscellγ -retroviraltransduction. (α-RVVs) vector In systemscontrast, are α-retroviral frequently vector used insystems gene therapy, have not particularly yet been forbroadly T- cellemployed transduction.γ-retroviral for stable vectorIn contrast, clinical systems gene α-retroviral are transfer. frequently vectorHowever, used systems in based gene have therapy,on their not particularly favorableyet been integrationbroadly for T-cell employedtransduction.profile and for thestable In possibility contrast, clinicalα gene -retroviralto establish transfer. vector permanent However, systems producerbased have on not celltheir yet lines, beenfavorable broadlyα-RVVs integration employedhave been profileforsuggested stable and clinicalthe as apossibility very gene promising transfer. to establish alternative However, permanent based to LVVs onproducer and their γ favorable-RVVs cell lines, [24,25,28]. integration α-RVVs In thehave proﬁle next been andpart suggested as a very promising alternative to LVVs and γ-RVVs [24,25,28]. In the next part

Viruses 2021, 13, 1471 6 of 11

the possibility to establish permanent producer cell lines, α-RVVs have been suggested as a very promising alternative to LVVs and γ-RVVs [24,25,28]. In the next part of the project, we thus addressed the possibility of using the coGALV-Env to pseudotype both

Virusesγ-RVVs 2021, 13, 1471 and α-RVVs. 6 of 11 To determine whether codon optimization has the same effect on the production of γ- RVVs and α-RVVs as on LVVs, a new vector preparation was conducted. Generally, based on preliminary resultsof the and project, published we thus addressed data on the optimal possibilityplasmid of using the concentrations coGALV-Env to pseudotype [24,29] we both γ-RVVs and α-RVVs. used the same experimentalTo determine setup whether for vector codon preparation optimization ha ass the shown same effect for on LVVs, the production i.e., multiple of parallel vector productionsγ-RVVs and with α-RVVs increasing as on LVVs, amounts a new vector of GALV-envpreparation was and conducted. coGALV-env Generally, and three times for harvestbased of supernatants.on preliminary results During and published vector production, data on optimal it wasplasmid interesting concentrations to note [24,29] we used the same experimental setup for vector preparation as shown for LVVs, that the 293T cells producingi.e., multipleα parallel-RVVs vector were productions less attached with increasing to the surface amounts of of the GALV-env tissue cultureand dish than those cellscoGALV-env producing and γthree-RVVs. times for Unexpectedly, harvest of supernatants. for both Duringγ-RVVs vector production, and α-RVVs, it vector productionswas with interesting standard to note GALV-env that the 293T did cells not producing reveal α-RVVs distinct wereoptima less attached of plasmidto the surface of the tissue culture dish than those cells producing γ-RVVs. Unexpectedly, for amounts, but titrationboth curves γ-RVVs showedand α-RVVs, rather vector wave-like productions shapes.with standard Nevertheless, GALV-env did vectornot reveal titers were in the expecteddistinct range optima for of bothplasmid vector amounts, types but titr (10ation7 curvestransducing showed rather units wave-like [TU]/mL shapes. for 7 γ-RVVs and 106 TU/mLNevertheless, for α -RVVs),vector titers independent were in the expected of the usedrange envfor both plasmid vector types (GALV-env (10 transducing units [TU]/mL for γ-RVVs and 106 TU/mL for α-RVVs), independent of the or coGALV-env). Mostused importantly,env plasmid (GALV-env for both or vectorcoGALV-env systems,). Most maximalimportantly, titers for both were vector again obtained with lowersystems, (approximately maximal titers 8–10 were times) again obtained plasmid with amounts lower (approximately of coGALV-env 8–10 times) (20 ng 6 plasmid amounts of coGALV-env (20 ng of plasmid/106 293T cells) as compared to of plasmid/10 293Tstandard cells) asGALV-env compared (Figure to 3).standard GALV-env (Figure3).

(a) (b) Figure 3. Titers of (a) γ- and (b) α-retroviral vectors produced with GALV-env or coGALV-env as determined on 293T Figurecells. For 3. transientTiters vector of ( aproduction) γ- and, the (b following) α-retroviral plasmid amounts vectors per produced 106 producer withcells were GALV-env used: GALV-env: or coGALV-env 1 µg, as0.64 determined µg, 0.32 µg, 0.16 on µg 293T and 0.08 cells. µg, coGALV-env: For transient 0.16 µg, vector 0.08 µg, production, 40 ng, 20 ng, 10 the ng, followingand 5 ng of plasmid. plasmid Data amountspoints per shown represent means of triplicates. SN = supernatant, TU = transducing units. 106 producer cells were used: GALV-env: 1 µg, 0.64 µg, 0.32 µg, 0.16 µg and 0.08 µg, coGALV-env: 0.16 µg, 0.08 µg, 40 ng, 20In ng, order 10 to ng, address and 5the ng potential of plasmid. impact of Data the internal points promotor shown representon vector titer, means the of triplicates. SN = supernatant,original EFS TU promotor = transducing in the α-RV units. was replaced by a stronger MPSV promotor, and a new vector production was performed for both γ-RVVs and α-RVVs following the same scheme as in Figure 3. Interestingly, very similar titration curves with no clear titer In order to addressmaxima the were potential found for impactboth vector of types the internal(not shown). promotor These data onmight vector indicate titer, that the original EFS promotorrelatively in the broadα-RV ranges was of replaced env plasmid by am aounts stronger around MPSV the determined promotor, maxima and still a new vector production wasfacilitate performed efficient vector for both production.γ-RVVs and α-RVVs following the same scheme Based on these data, we used the following plasmid amounts (per 106 producer cells) as in Figure3. Interestingly,for standard-scale very similar manufacture titration of both vector curves types with (γ-RVVs no clear and α-RVVs)—GALV-env: titer maxima were found for both vector160 types ng, coGALV-env: (not shown). 20 ng. The These supernatants data mightproduced indicate in the standard-scale that relatively runs for the broad ranges of env plasmid amounts around the determined maxima still facilitate efﬁcient vector production.

Based on these data, we used the following plasmid amounts (per 106 producer cells) for standard-scale manufacture of both vector types (γ-RVVs and α-RVVs)—GALV-env: 160 ng, coGALV-env: 20 ng. The supernatants produced in the standard-scale runs for the two vectors were titrated on 293T cells as above; results are displayed in Figure4. As evident, we again obtained very similar titers for both vector constructs, despite the fact that we used eight times lower amounts of the plasmid encoding coGALV-Env. Viruses 2021, 13, 1471 7 of 11

Viruses 2021, 13, 1471 7 of 11 two vectors were titrated on 293T cells as above; results are displayed in Figure 4. As evident, we again obtained very similar titers for both vector constructs, despite the fact that we used eight times lower amounts of the plasmid encoding coGALV-Env.

(a) (b)

γ α FigureFigure 4. Titers 4. Titers of vectorsof vectors produced produced in standard-scalein standard-scale with with GALV-env GALV-env and and coGALV-env coGALV-env for for -RVsγ-RVs and and -RVs.α-RVs. GALV-env- GALV- 6 6 vectorsenv-vectors were produced were produced with 160 with ng 160 of plasmid/10ng of plasmid/10cells,6 cells, coGALV-env-vectors coGALV-env-vectors with with 20 20 ng ng of of plasmid/10 plasmid/106cells. cells. TitersTiters of (a) γof-RVVs (a) γ-RVVs and (b and) α-RVVs (b) α-RVVs were determinedwere determined on 293T on 293T cells. cells. Data Data shown shown are the are mean the mean of triplicates. of triplicates. TU =TU transducing = transducing units, * p

3.3. Transduction of K562 Cells and Primary Human T Lymphocytes with All Three Vector 3.3.Types Transduction of K562 Cells and Primary Human T Lymphocytes with All Three Vector Types InIn the the ﬁnal final set ofset experiments, of experiments, we used we allused three all vectorthree typesvector to types transduce to transduce hematopoi- etichematopoietic cells, namely, cells, the namely, human the erythro-leukemia human erythro-leukemia cell line K562, cell line and, K562, more and, importantly, more primaryimportantly, human primary T lymphocytes. human T lymphocytes. K562 cells were K562 chosen cells aswere a broadly chosen usedas a broadly modelcell used line, whereasmodel cell primary line, whereas T cells representprimary T a cells highly represent relevant a highly target forrelevant genetic target modiﬁcations for genetic in immunotherapy.modifications in Moreover,immunotherapy. all three Moreover, vector types all three have vector been successfullytypes have been used successfully to transduce primaryused to Ttransduce cells previously. primary T cells previously. InIn order order to to directly directly assess assess the the potential potential impact impact ofof thethe GALV-envGALV-env plasmidplasmid (standard(standard vs. codon-optimized),vs. codon-optimized), we determined we determined titers titers of the of the vector vector preparations preparations from from the the standard-scale standard- productionsscale productions for all threefor all vectorthree vector types types (LVVs, (LVVs,γ-RVVs, γ-RVVs, and αand-RVVs). α-RVVs). To thisTo this end, end, we we used standardused standard dilution dilution series asseries described as described above. abov Cell-typee. Cell-type specific specific titers titers of the of differentthe different vector preparationsvector preparations are depicted are depicted in Figure in5, Figure separately 5, separately for K562 (Figurefor K5625a) (Figure and T 5a) cells and (Figure T cells5b). As(Figure evident, 5b). only As minorevident, differences only minor between differences the two between envelopes the two without envelopes clear tendency without inclear favor oftendency one or the in other favor were of one found or the for other the three were vector found types. for the In particular,three vector in primarytypes. Inhuman particular, T cells, weinobserved primary slightlyhuman (factorT cells, 1.8) we higher observed titers forslightly LVV pseudotyped(factor 1.8) higher with the titers standard for LVV GALV- Env,pseudotyped while for both withγ the-RVVs standard and α GALV-Env,-RVVs the titers while were for both slightly γ-RVVs higher and with α-RVVs the coGALV-Env the titers (bywere factors slightly 1.5 and higher 1.3, with respectively). the coGALV-Env Although (by these factors minor 1.5 and variances 1.3, respectively). between GALV-env Although and these minor variances between GALV-env and coGALV-env would not be expected to be coGALV-env would not be expected to be relevant in most cases, our data confirmed the notion relevant in most cases, our data confirmed the notion that in the direct comparison of the that in the direct comparison of the three vectors, GALV-env pseudotyped γ-RVVs might be three vectors, GALV-env pseudotyped γ-RVVs might be most useful for primary human most useful for primary human T cells. T cells.

VirusesViruses 20212021,, 1313,, 14711471 88 of of 11

(a) (b)

Figure 5. Titers of standard-scale produced supernatants of the three vector types (for LVVs, γ-RVVs, and α-RVVs) on Figure 5. Titers of standard-scale produced supernatants of the three vector types (for LVVs, γ-RVVs, and α-RVVs) on K562 cellscells andand primaryprimary humanhuman T T lymphocytes. lymphocytes. LVVs LVVs were were produced produced with with 640 640 ng ng and andγ-RVV γ-RVV as wellas well as αas-RVV α-RVV with with 160 160 ng 6 6 GALV-envng GALV-env plasmids plasmids per 10 perproducer 106 producer cells. In cells. contrast, In contrast, for production for production of all coGALV-env of all coGALV-env vectors, only vectors, 20 ng ofonly plasmid/10 20 ng of cellsplasmid/10 was used.6 cells Titers was were used. determined Titers were using determined dilution seriesusing ondilution (a) K562 series and on (b) ( primarya) K562 Tand lymphocytes. (b) primary Shown T lymphocytes. are means ofShown triplicates. are means TU = of transducing triplicates. units,TU = transducing * p < 0.01 (t-test). units, * p < 0.01 (t-test).

4. Discussion 4. Discussion Based on their ability to transduce human hematopoietic cells with minimal cytotoxicity,Based on GALV-Env-pseudotyped their ability to transduce humanretroviral hematopoietic vectors are cellsof great with interest minimal for cytotoxic- various ity, GALV-Env-pseudotyped retroviral vectors are of great interest for various gene-therapy gene-therapy applications. Indeed, already in the 1990s it was shown for γ-RVVs that the applications. Indeed, already in the 1990s it was shown for γ-RVVs that the GALV-Env GALV-Env protein mediates efficient transduction of primary T lymphocytes [9,11]. protein mediates efficient transduction of primary T lymphocytes [9,11]. Subsequently, Subsequently, the GALV-Env protein was modified to allow pseudotyping of LVVs [12– the GALV-Env protein was modified to allow pseudotyping of LVVs [12–15]. More re- 15]. More recently, pseudotyping of α-RVVs was also achieved but resulted in cently, pseudotyping of α-RVVs was also achieved but resulted in comparatively low comparatively low vector titers [23,24]. vector titers [23,24]. In view of the recent progress in adoptive immunotherapy with T cells expressing In view of the recent progress in adoptive immunotherapy with T cells expressing tumor-targeting TCRs [30] and, particularly, CARs [16], efficient gene transfer into T tumor-targeting TCRs [30] and, particularly, CARs [16], efficient gene transfer into T lym- lymphocytes has become a very urgent task. Here, we asked whether codon-optimization phocytes has become a very urgent task. Here, we asked whether codon-optimization of of the GALV-env ORF might contribute to improved vector titers (in particular for α- the GALV-env ORF might contribute to improved vector titers (in particular for α-RVVs) RVVs) and the reduction of plasmid amounts required for production of all vector types. and the reduction of plasmid amounts required for production of all vector types. The lat- terThe point latter will point be relevantwill be relevant in the context in the ofcontext manufacturing of manufacturing schemes basedschemes on based transient on transfectiontransient transfection of producer of cell producer lines where cell largelines amountswhere large of plasmid amounts of goodof plasmid manufacturing of good practicemanufacturing (GMP) practice quality are(GMP) required. quality Notably, are required. manufacturing Notably, manufacturing costs of cell-therapy costs of drugs cell- stilltherapy represent drugs anstill important represent limitation an important for broader limitation application for broader [31 ].application [31]. For γ-RVVs and LVVs, wewe observedobserved that application of the coGALV-Env duringduring transient vectorvector production production resulted resulted in very in very similar similar titers astiters the originalas the plasmid.original However,plasmid. theHowever, amount the of amount necessary of GALV-envnecessary GALV-env plasmid could plasmid be reduced could be by reduced a factor by of >30a factor (20 ngof instead>30 (20 ng of instead 640 ng perof 640 1 Mio ng per producer 1 Mio producer cells) for LVVscells) for and LVVs by a factorand by of a eightfactor forof eightγ-RVVs for (20γ-RVVs ng instead (20 ng of instead 160 ng/10 of 1606 cells). ng/10 Notably,6 cells). theNotably, applied the quantity applied of quantity just 20 ng of ofjust packaging 20 ng of plasmidpackaging for plasmid coGALV-env for coGALV-env is substantially is substantially lower than thelower amounts than the used amounts by many used other by groupsmany other to pseudotype groups toγ pseudotype-RVVs and LVVs γ-RVVs [12 ,13and,25 LVVs,26,28 ].[12,13,25,26,28]. We also obtained We relatively also obtained good titersrelatively (>10 6goodtransducing titers (>10 units6 transducing per ml of non-concentratedunits per ml of no supernatant)n-concentrated for supernatant)α-RVVs, which for isα-RVVs, somewhat which in contrastis somewhat to previous in contrast work [to23 previous,24]. Again, work the [23,24]. necessary Again, amount the of necessary plasmid wasamount strongly of plasmid reduced was for strongly the codon-optimized reduced for the codon-optimized construct (20 ng construct vs. 160ng (20 per ng vs. 1 Mio 160 producerng per 1 Mio cells) producer and >10 cells) times and lower >10 than times previously lower than reported previously [23]. reported [23]. Codon optimization was previously used to optimizeoptimize productionproduction ofof pseudotypedpseudotyped LVVs and γ-RVVs,-RVVs, e.g.,e.g., withwith SpikeSpike proteinprotein fromfrom SARS-CoV-1SARS-CoV-1 [[32],32], parainfluenzaparainfluenza typetype 33 envelope proteins [33], [33], and a truncated HIV-Env [[34].34]. In striking contrast, Zucchelli et et al. al. observed impaired functionality of codon-optimizedcodon-optimized RD114-TR, another envelope protein derived from a γ-retrovirus thatthat isis commonlycommonly usedused forfor pseudotypingpseudotyping [[20].20]. The latter data have thus proven that codon optimization of the env genegene doesdoes notnot necessarilynecessarily improveimprove protein expression or vectorvector titers.titers.

Viruses 2021, 13, 1471 9 of 11

In our work we did not see any functional impairment for any of the three tested vector types (LVVs, γ-RVVs, and α-RVVs) produced with the coGALV-env. In direct comparison, coGALV-Env pseudotyped γ-RVVs showed the highest titers (approximately 1 × 107 TU per mL on 293T cells). This was to be expected, since GALV-Env is derived from a γ-RVV and had to be modiﬁed (chimeric GALV-Env) to be efﬁciently incorporated into the other vector particles [12,13,23]. With regard to practical use, the observed high transduction rates on primary T lymphocytes with coGALV-Env γ-RVVs might be of particular interest.

5. Conclusions The coGALV-Env facilitates efﬁcient pseudotyping of lentiviral, γ-retroviral, and α- retroviral vectors. For all three vector systems, signiﬁcantly lower amounts of plasmid of coGALV-env are required to gain identical titers and transduction rates, as compared with classical GALV-env. These data might be of particular relevance for large-scale, GMP vector production using transient producer cell lines.

Supplementary Materials: The following SM is available online at https://www.mdpi.com/article/ 10.3390/v13081471/s1, Figure S1: Comparison of the complete original and the optimized sequences of GALV-env. Figure S2: Vector maps of both envelope expressing plasmids used in this work. Figure S3: Dose-finding of coGALV-env plasmid amount needed for efficient packaging of lentiviral vectors. Author Contributions: Conceptualization, B.F. and K.R.; methodology, M.M. and L.I.S.; experimental work, M.M., L.I.S. and K.R.; writing—original draft preparation, B.F.; writing—data visualization, review, and editing, M.M. and K.R. All authors have read and agreed to the published version of the manuscript. Funding: This research was partly supported by Deutsche Forschungsgemeinschaft (SFB841, SP2). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Acknowledgments: The authors wish to thank Tanja Sonntag and Almut Uhde for expert technical assistance. We are also indebted to the members of the UKE Cytometry and Cell Sorting Core facility and the Institute of Transfusion Medicine for their kind support. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References 1. Blaese, R.M.; Culver, K.W.; Miller, A.D.; Carter, C.S.; Fleisher, T.; Clerici, M.; Shearer, G.; Chang, L.; Chiang, Y.; Tolstoshev, P.; et al. T lymphocyte-directed gene therapy for ADA-SCID: Initial trial results after 4 years. Science 1995, 270, 475–480. [CrossRef] [PubMed] 2. Cartier, N.; Hacein-Bey-Abina, S.; Bartholomae, C.C.; Veres, G.; Schmidt, M.; Kutschera, I.; Vidaud, M.; Abel, U.; Dal-Cortivo, L.; Caccavelli, L.; et al. Hematopoietic stem cell gene therapy with a lentiviral vector in X-linked adrenoleukodystrophy. Science 2009, 326, 818–823. [CrossRef] 3. Morgan, M.A.; Galla, M.; Grez, M.; Fehse, B.; Schambach, A. Retroviral gene therapy in Germany with a view on previous experience and future perspectives. Gene Ther. 2021. prepublished online. [CrossRef] 4. Miller, A.D.; Garcia, J.V.; von Suhr, N.; Lynch, C.M.; Wilson, C.; Eiden, M.V. Construction and properties of retrovirus packaging cells based on gibbon ape leukemia virus. J. Virol. 1991, 65, 2220–2224. [CrossRef] 5. Takeuchi, Y.; Simpson, G.; Vile, R.G.; Weiss, R.A.; Collins, M.K. Retroviral pseudotypes produced by rescue of a Moloney murine leukemia virus vector by C-type, but not D-type, retroviruses. Virology 1992, 186, 792–794. [CrossRef] 6. Burns, J.C.; Friedmann, T.; Driever, W.; Burrascano, M.; Yee, J.K. Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: Concentration to very high titer and efﬁcient gene transfer into mammalian and nonmammalian cells. Proc. Natl. Acad. Sci. USA 1993, 90, 8033–8037. [CrossRef] 7. Joglekar, A.V.; Sandoval, S. Pseudotyped Lentiviral Vectors: One Vector, Many Guises. Hum. Gene Ther. Methods 2017, 28, 291–301. [CrossRef] 8. Miller, A.D.; Chen, F. Retrovirus packaging cells based on 10A1 murine leukemia virus for production of vectors that use multiple receptors for cell entry. J. Virol. 1996, 70, 5564–5571. [CrossRef] Viruses 2021, 13, 1471 10 of 11

9. Bunnell, B.A.; Muul, L.M.; Donahue, R.E.; Blaese, R.M.; Morgan, R.A. High-efficiency retroviral-mediated gene transfer into human and nonhuman primate peripheral blood lymphocytes. Proc. Natl. Acad. Sci. USA 1995, 92, 7739–7743. [CrossRef] [PubMed] 10. Girard-Gagnepain, A.; Amirache, F.; Costa, C.; Lévy, C.; Frecha, C.; Fusil, F.; Nègre, D.; Lavillette, D.; Cosset, F.L.; Verhoeyen, E. Baboon envelope pseudotyped LVs outperform VSV-G-LVs for gene transfer into early-cytokine-stimulated and resting HSCs. Blood 2014, 124, 1221–1231. [CrossRef][PubMed] 11. Ayuk, F.; Li, Z.; Kühlcke, K.; Lindemann, C.; Schade, U.; Eckert, H.G.; Zander, A.; Fehse, B. Establishment of an optimised gene transfer protocol for human primary T lymphocytes according to clinical requirements. Gene Ther. 1999, 6, 1788–1792. [CrossRef] [PubMed] 12. Stitz, J.; Buchholz, C.J.; Engelstadter, M.; Uckert, W.; Bloemer, U.; Schmitt, I.; Cichutek, K. Lentiviral vectors pseudotyped with envelope glycoproteins derived from gibbon ape leukemia virus and murine leukemia virus 10A1. Virology 2000, 273, 16–20. [CrossRef][PubMed] 13. Christodoulopoulos, I.; Cannon, P.M. Sequences in the cytoplasmic tail of the gibbon ape leukemia virus envelope protein that prevent its incorporation into lentivirus vectors. J. Virol. 2001, 75, 4129–4138. [CrossRef][PubMed] 14. Mock, U.; Thiele, R.; Uhde, A.; Fehse, B.; Horn, S. Efficient lentiviral transduction and transgene expression in primary human B cells. Hum. Gene Ther. Methods 2012, 23, 408–415. [CrossRef][PubMed] 15. Berdien, B.; Mock, U.; Atanackovic, D.; Fehse, B. TALEN-mediated editing of endogenous T-cell receptors facilitates efficient reprogramming of T lymphocytes by lentiviral gene transfer. Gene Ther. 2014, 21, 539–548. [CrossRef][PubMed] 16. June, C.H.; Sadelain, M. Chimeric Antigen Receptor Therapy. N. Engl. J. Med. 2018, 379, 64–73. [CrossRef] 17. Ferreira, M.V.; Cabral, E.T.; Coroadinha, A.S. Progress and Perspectives in the Development of Lentiviral Vector Producer Cells. Biotechnol. J. 2021, 16, e2000017. [CrossRef][PubMed] 18. Wagner, R.; Graf, M.; Bieler, K.; Wolf, H.; Grunwald, T.; Foley, P.; Uberla, K. Rev-independent expression of synthetic gag-pol genes of human immunodeficiency virus type 1 and simian immunodeficiency virus: Implications for the safety of lentiviral vectors. Hum. Gene Ther. 2000, 11, 2403–2413. [CrossRef] 19. Kotsopoulou, E.; Kim, V.N.; Kingsman, A.J.; Kingsman, S.M.; Mitrophanous, K.A. A Rev-independent human immunodeficiency virus type 1 (HIV-1)-based vector that exploits a codon-optimized HIV-1 gag-pol gene. J. Virol. 2000, 74, 4839–4852. [CrossRef] 20. Zucchelli, E.; Pema, M.; Stornaiuolo, A.; Piovan, C.; Scavullo, C.; Giuliani, E.; Bossi, S.; Corna, S.; Asperti, C.; Bordignon, C.; et al. Codon Optimization Leads to Functional Impairment of RD114-TR Envelope Glycoprotein. Mol. Ther. Methods Clin. Dev. 2017, 4, 102–114. [CrossRef] 21. Mohme, M.; Maire, C.L.; Riecken, K.; Zapf, S.; Aranyossy, T.; Westphal, M.; Lamszus, K.; Fehse, B. Optical Barcoding for Single-Clone Tracking to Study Tumor Heterogeneity. Mol. Ther. 2017, 25, 621–633. [CrossRef] 22. Weber, K.; Bartsch, U.; Stocking, C.; Fehse, B. A multicolor panel of novel lentiviral “gene ontology” (LeGO) vectors for functional gene analysis. Mol. Ther. 2008, 16, 698–706. [CrossRef][PubMed] 23. Schambach, A.; Mueller, D.; Galla, M.; Verstegen, M.M.; Wagemaker, G.; Loew, R.; Baum, C.; Bohne, J. Overcoming promoter competition in packaging cells improves production of self-inactivating retroviral vectors. Gene Ther. 2006, 13, 1524–1533. [CrossRef] 24. Suerth, J.D.; Maetzig, T.; Galla, M.; Baum, C.; Schambach, A. Self-inactivating alpharetroviral vectors with a split-packaging design. J. Virol. 2010, 84, 6626–6635. [CrossRef][PubMed] 25. Labenski, V.; Suerth, J.D.; Barczak, E.; Heckl, D.; Levy, C.; Bernadin, O.; Charpentier, E.; Williams, D.A.; Fehse, B.; Verhoeyen, E.; et al. Alpharetroviral self-inactivating vectors produced by a superinfection-resistant stable packaging cell line allow genetic modification of primary human T lymphocytes. Biomaterials 2016, 97, 97–109. [CrossRef][PubMed] 26. Tomás, H.A.; Mestre, D.A.; Rodrigues, A.F.; Guerreiro, M.R.; Carrondo, M.J.T.; Coroadinha, A.S. Improved GaLV-TR Glycoproteins to Pseudotype Lentiviral Vectors: Impact of Viral Protease Activity in the Production of LV Pseudotypes. Mol. Ther. Methods Clin. Dev. 2019, 15, 1–8. [CrossRef][PubMed] 27. Dautzenberg, I.J.C.; Rabelink, M.J.W.E.; Hoeben, R.C. The stability of envelope-pseudotyped lentiviral vectors. Gene Ther. 2021, 28, 89–104. [CrossRef] 28. Moiani, A.; Suerth, J.D.; Gandolfi, F.; Rizzi, E.; Severgnini, M.; De Bellis, G.; Schambach, A.; Mavilio, F. Genome-wide analysis of alpharetroviral integration in human hematopoietic stem/progenitor cells. Genes 2014, 5, 415–429. [CrossRef] 29. van der Loo, J.C.; Swaney, W.P.; Grassman, E.; Terwilliger, A.; Higashimoto, T.; Schambach, A.; Baum, C.; Thrasher, A.J.; Williams, D.A.; Nordling, D.L.; et al. Scale-up and manufacturing of clinical-grade self-inactivating γ-retroviral vectors by transient transfection. Gene Ther. 2012, 19, 246–254. [CrossRef] 30. Manfredi, F.; Cianciotti, B.C.; Potenza, A.; Tassi, E.; Noviello, M.; Biondi, A.; Ciceri, F.; Bonini, C.; Ruggiero, E. TCR Redirected T Cells for Cancer Treatment: Achievements, Hurdles, and Goals. Front. Immunol. 2020, 11, 1689. [CrossRef][PubMed] 31. Heine, R.; Thielen, F.W.; Koopmanschap, M.; Kersten, M.J.; Einsele, H.; Jaeger, U.; Sonneveld, P.; Sierra, J.; Smand, C.; Uyl-de Groot, C.A. Health Economic Aspects of Chimeric Antigen Receptor T-cell Therapies for Hematological Cancers: Present and Future. Hemasphere 2021, 5, e524. [CrossRef][PubMed] 32. Moore, M.J.; Dorfman, T.; Li, W.; Wong, S.K.; Li, Y.; Kuhn, J.H.; Coderre, J.; Vasilieva, N.; Han, Z.; Greenough, T.C.; et al. Retroviruses pseudotyped with the severe acute respiratory syndrome coronavirus spike protein efficiently infect cells expressing angiotensin-converting enzyme 2. J. Virol. 2004, 78, 10628–10635. [CrossRef][PubMed] Viruses 2021, 13, 1471 11 of 11

33. Jung, C.; Le Doux, J.M. Lentiviruses inefﬁciently incorporate human parainﬂuenza type 3 envelope proteins. Biotechnol. Bioeng. 2008, 99, 1016–1027. [CrossRef] 34. Siegert, S.; Thaler, S.; Wagner, R.; Schnierle, B.S. Assessment of HIV-1 entry inhibitors by MLV/HIV-1 pseudotyped vectors. AIDS Res. Ther. 2005, 2, 7. [CrossRef][PubMed]